The Sport Motivational Scale (SMS; Pelletier, et. al., 1995) was developed to assess motivation at the contextual level. Subsequently, but not without criticisms, the SMS has been used to assess contextual motivation in Physical Education (PE) in both cross-sectional and experimental designs. As a result, various versions of the SMS have been used, or have been proposed for use, but not necessarily having established the validity of each new version. The purpose of this study was to assess the psychometric properties and tenability of the SMS, in its various actual and proposed forms, within the context of PE via confirmatory factor analysis (CFA).
Method:
SMS data were collected from 1238 male and female Jr. High PE students in the US and UK near the end of the school year. Correlational analysis and internal consistency (Chronbach's alpha) were used to assess the proposed SIMPLEX structure of the SMS. CFA was used to assess fit indices for several versions of the SMS used in the PE Setting. Each porposed version of the SMS were for goodness of fit and then compared for suitability to the PE setting.
Analysis/Results:
Reliability tests indicated adequate internal consistencies and support for the SIMPLEX pattern of associations proposed by self-determination theory (SDT). CFA results comparing several full and reduced models supported (a) that the SMS was invariant across gender and location, (b) removing five items (10, 11, 17, 26, and 28) of the original 28 items markedly improved goodness-of-fit, and (c) that alternatives to the original 7-factor, 28-items scale (i.e., 7-factor, 23-item; 6-factor, 23-item; 4-factor, 12- and 23-item models) each demonstrate marked improvement of fit indices.
Conclusions:
First, with minor modifications and with careful theoretical considertaions discussed herein, the SMS remains a practical and viable measurement tool for use in the PE setting. Second, if shorter versions are deemed necessary or desireable, two reduced SMS models are recommended for certain uses.